View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Research Papers in Economics The Pakistan Development Review 44 : 4 Part II (Winter 2005) pp. 1111–1131 Prevalence of Relative Poverty in Pakistan TALAT ANWAR* I. INTRODUCTION Much has been written11about poverty in Pakistan. A large number of attempts have been made by various authors/institutions to estimate the poverty in Pakistan over the last four decades. However, the conceptual basis of poverty remained limited to absolute concept of poverty. The concept of absolute poverty emphasises to estimate the cost of purchasing a minimum ‘basket’ of goods required for human survival. In Pakistan, the discussion has been centered on estimating poverty lines consistent with 2550 or 2350 calorie intake per adult per day as minimum requirement. Thus, absolute definitions of poverty tend to be minimalist and are based on subsistence and the attainment of physical efficiency. Subsistence is concerned with the minimum provision needed to maintain health and working capacity. However, the concept of absolute poverty has been criticised2 on the grounds that it minimises the range and depth of human needs. Human needs are interpreted as predominantly physical needs rather than social needs. People are relatively deprived if they cannot take part in the ordinary way of life of the community and cannot play their roles by virtue of their membership of the society. Furthermore, there have been difficulties in substantiating the absolute poverty approaches in robust empirical terms. This led analysts to a social formulation of the meaning of poverty—relative deprivation which some have defined as having income less than Talat Anwar is Senior Economist at the UNDP/UNOPS Project—Centre for Research on Poverty Reduction and Income Distribution, Islamabad. Author’s Note: Technical help for HIES data from Mr Hammad Ali, Federal Bureau of Statistics, Islamabad, is gratefully acknowledged. The views expressed are those of the author and do not necessarily reflect those of the CRPRID, UNDP/UNOPS, or the Planning Commission. 1See Naseem (1973, 1979), Allaudin (1975), Mujahid (1978), Irfan and Amjad (1984), Malik. (1988), Malik (1991), Anwar (1996), FBS(2001), World Bank (2002), Arif (2002), Anwar and Qureshi (2003), Planning Commission (2003) and Anwar, Qureshi, and Ali (2005). 2See Rein (1970) and Townsend (1970). 1112 Talat Anwar half or two-third or three-fourth of average expenditure (or income) norm [Rein (1970) and Townsend (1970)]. The concept of relative poverty has gained more importance because of recent rising trends in inequality developing countries and its linkages with poverty. For a given mean income, the more unequal the income distribution, the larger the percentage of the population living in poverty. In this context, the objective of this paper is to broaden the discussion on poverty and poverty measurement by examining the prevalence of relative poverty in Pakistan using the most recent available household survey data—HIES, 2000-02. The organisation of the paper is follows: The next section provides a review of assessment of poverty. Section III discusses the methods of measurement of poverty. The data set of Household Integrated Economic Survey (HIES), 2001-02 that has been used to examine the prevalence of relative poverty is discussed in Section IV. Section V presents the results for the prevalence of relative poverty in Pakistan. Section VI analyses the trends in relative poverty over time. Section VII discusses trends in income inequality over time. Main conclusions and policy implications conclude the discussion in the final section. II. REVIEW OF POVERTY ASSESSMENT IN PAKISTAN A large number of attempts have been made to estimate incidence of poverty in Pakistan during the last four decades. The earliest work on poverty was pioneered by Naseem (1973) by choosing an arbitrary poverty line to estimate the poverty incidence in Pakistan. Various attempts on poverty include, Naseem (1973, 1979), Alauddin (1975), Mujahid (1978), Amjad and Irfan (1984), Kruik and Leeuwen (1985), Malik (1988), Havinga, et al. (1989), Ahmad and Ludlow (1989), Ercelawn (1992), Malik (1992), Zaidi (1992) and Malik (1992), Anwar (1996, 1996a), Anwar (1998), FBS (2001), World Bank (2002, 2005), Arif (2002), Anwar and Quershi (2003), Planning Commission (2003), SPDC (2005) and Anwar, Qureshi, and Ali (2005). However, except few studies most of the work is based on the absolute concept of poverty. Various authors/institutions employed different methods, defined poverty in different ways and chose different poverty lines in absolute term, and thereby have reported divergent poverty trends. A general consensus emerging from the literature is that absolute poverty has declined during the 1970s and 1980s. The decline in absolute poverty was mainly attributable to the high economic growth as the economy witnessed an average growth rate at about 5 percent and 6 percent per annum respectively, in the 1970s and 1980s mainly due heavy capital inflows from abroad in the forms of foreign aid and overseas workers remittances. However, in contrast to the 1970s and 1980s, absolute poverty rose during the 1990s. Except the World Bank (2002) that concludes a stagnant level of absolute poverty, there is a general consensus in the literature of rising levels of absolute poverty in Pakistan during the 1990s [FBS Prevalence of Relative Poverty 1113 (2001), World Bank (2002, 2005), Anwar and Quershi (2003), Planning Commission (2003) and Anwar, Qureshi, and Ali (2005)]. A review of poverty assessment shows that while there has been a heated debated on the levels of absolute poverty during the 1990s, the levels of relative poverty from the viewpoint of policy reforms has not received adequate attention. There are, nonetheless, two studies, Zaidi (1992) and Anwar (1998) that examined the extent of relative poverty in Pakistan. While measuring relative poverty, Zaidi (1992) took 75 percent of the national average expenditure as relative poverty line using the HIES data for 1984-85. The monthly average expenditure was at Rs 340 implying that the poverty line was at Rs 255 per capita expenditure in 1984-85 prices. The author used equivalence scale recommended by OECD [OECD (1982)]. This scale assigns a weight of one to the first adult and, 0.7 to every other adult and 0.5 to children of age 13 years and younger. The poverty line was assumed to be identical across the four provinces and rural and urban regions. The author found that almost 39 percent of households are below the poverty line in Pakistan in 1984-85. The poverty headcount were found to be highly sensitive to the choice of percentage of the average to express poverty line. The headcount was only 15 percent when the poverty line was fixed at 50 percent of national average expenditure. Provincial estimates suggest that Sindh and the NWFP were the least poor whereas Punjab and Balochistan were the poorest provinces in Pakistan in 1984-85. It may be pointed out that poverty ranking across provinces implied by Zaidi (1992) appears to be contrary to general the perception as well as the evidence supported by other authors. This contradictory ranking is due to the fact that the author used identical poverty lines for all the provinces which appear to be misleading because of the possible differences in the cost of living across provinces and regions. Thus, adequate differences in the cost of living between rural and urban should be taken into account to estimate the poverty across provinces. Anwar (1998) chose three cut-off points at 50 percent, 66.6 percent and 75 percent of national adult equivalent consumption expenditure to define the relative poverty line in 1987-88. To correct the consumption expenditure data for household size composition, the author used his derived equivalence scales based on Engel’s method. The scale suggests a first male child costing 86 percent of an adult, a second male child costing 78 percent of an adult, whereas first female child costing 75 percent and second female child costing 61 percent of an adult in Pakistan. The author derived relative poverty threshold as 50 percent, 66.6 percent and 75 percent of national adult equivalent expenditure which for overall Pakistan turned out to be at Rs 251, Rs 336 and Rs 380 adult equivalent expenditure, respectively. Using these cut-off points the author found that 14.7, 39.2 and 48.2 percent of all households were poor in 1987-88. The author used the region and province-specific relative poverty cut-off points to take an account of differences in the cost of living across provinces and regions while estimating the extent of relative poverty across regions 1114 Talat Anwar and provinces. This approach takes not only an account of food price differences but also of behavioural differences in cost of living across regions. While the average expenditure is much higher in urban areas than in rural areas, this approach is likely to derive significantly higher poverty line for urban areas compared to the rural areas leading to higher level of poverty in urban than in rural areas. It is, therefore, pertinent to take an account of differences only in food prices in poverty line between rural and urban areas. This is because cost of living is higher in urban than in rural areas due to differences in food prices. For example, if two households have exactly the same standard of living but reside in different regions, then consistency requires that poverty line be adjusted accordingly to the food price differences. For this reason, the paper takes an account of differences in food prices 3 between rural and urban and among provinces by using a regional price 3 index. It is turned out that on average food prices are 13 percent higher in urban areas than in rural areas.
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages21 Page
-
File Size-